The MRP family of drug efflux pumps


The MRP family is comprised of nine related ABC transporters that are able to transport structurally diverse lipophilic anions and function as drug efflux pumps. Investigations of this family have provided insights not only into cellular resistance mechanisms associated with natural product chemotherapeutic agents, antifolates and nucleotide analogs, but also into factors that influence drug distribution in the body, membrane systems that are involved in the extrusion of reduced folates, cysteinyl leukotrienes and bile acids, and the molecular basis of two hereditary conditions in humans. The review will describe the biochemical properties, drug resistance activities and potential in vivo functions of these unusual pumps.


Cell lines made resistant to anticancer agents have served as important tools in the elucidation of cellular resistance mechanisms. This is well illustrated by the identification and characterization of the founding member of the MRP family. Investigations of HL60/ADR, a particularly informative doxorubicin-resistant cell line, revealed that it exhibited a multidrug resistance phenotype in association with an energy-dependent drug accumulation deficit, but did not overexpress P-glycoprotein (McGrath and Center, 1987; McGrath et al., 1989). Instead, the phenotype of HL60/ADR was attributable to the overexpression of a 190 kDa resistance-associated protein, which was determined to be an ABC transporter by virtue of its susceptibility to labeling by the nucleotide binding fold label AzATP, and its immunoreactivity with peptide antisera directed at the conserved nucleotide binding site of P-glycoprotein (McGrath et al., 1989; Marquardt et al., 1990). The isolation of the multidrug resistance-associated protein cDNA (now called MRP1) from another drug-resistant cell line (Cole et al., 1992) revealed the molecular identity of the 190 kDa protein (Krishnamachary and Center, 1993; Kruh et al., 1994), and provided the molecular tools that established an unexpected connection between resistance to anticancer agents and efflux systems responsible for extrusion of a remarkable range of amphipathic anions. This review will focus primarily on the drug resistance and biochemical activities of the MRP family, and their potential in vivo functions. Aspects of gene regulation are described in the review by Scotto, in this issue.

The MRP family

The amino-acid sequence of MRP1 resembles P-glycoprotein to only a modest extent (15%), and its structure is distinct as well. MRP1 is composed of a large ‘core’ segment that is similar to P-glycoprotein (MSD1 → C-terminus), but has in addition an N-terminal region that is composed of a third membrane spanning domain (MSD0) in which five transmembrane helices reside, an intracellular loop (L0) and an extracellular N-terminus (Figure 1) (Bakos et al., 1996; Hipfner et al., 1997; Kast and Gros, 1997). Thus, MRP1 can be thought of as having an N-terminal hydrophobic domain (MSD0) linked via a cytoplasmic loop (L0) to a Pgp-like core. The core region has two nucleotide binding domains (NBDs), two membrane spanning domains each composed of six transmembrane spanning helices and a linker segment (L1) located between NBD1 and MSD2. MRP1 is now known to be the founding member of a family of proteins that extends to nine members. Analyses of the predicted structures of subsequently identified MRPs indicated that not all of them possess MSD0 domains (Belinsky et al., 1998). MRP4, MRP5 and the recently identified MRP8 and MRP9 lack the third membrane spanning domain (MSD0), but possess L0 domains, whereas MRP2, MRP3, MRP6 and MRP7 resemble MRP1 (Bera et al., 2001; Hopper et al., 2001; Tammur et al., 2001; Yabuuchi et al., 2001) (Figure 1). Remarkably, all of the characterized members of this family are lipophilic anion pumps that have the facility for conferring resistance to anticancer agents. The properties of these pumps are summarized in Table 1 and Figure 2, and are discussed in detail below.

Figure 1

Topology of MRP family members. (a) Schematic depicting the organization of protein domains. Stripes, membrane spanning domains; open, cytoplasmic loops located between MSD0 and MSD1, NBF1 and MSD2 and at the C-terminus; black, nucleotide binding folds. (b) Topological model of MRP1 (which resembles MRP2, MRP3, MRP6 and MRP7) (top) and MRP4 (which resembles MRP5, MRP8 and MRP9) (bottom). NBF, nucleotide binding fold MSD, membrane spanning domain. MSD1 domains of the proteins are aligned (adapted from Hopper et al., 2001)

Table 1 Summary of MRP family members
Figure 2

Subcellular localization of MRPs in polarized epithelial cells. The localization of MRPs in epithelial cells surrounding a hypothetical lumen is shown. MRP1, MRP3, MRP5 and MRP6 are localized in basolateral membranes. MRP2 is localized in apical membranes. MRP4 is localized in basolateral membranes in human prostatic glandular cells and in apical membranes in rat kidney tubule cells. The localizations of MRP7, MRP8 and MRP9 have not been determined. Pgp and ABCG2 are apical efflux pumps (not shown)


Glutathione and glucuronate conjugate pump and resistance factor for anthracyclines, epipodophyllotoxins, vinca alkaloids and camptothecins

At the time of the molecular identification of MRP1, its modest degree of sequence similarity with Pgp was striking in view of the overlap of their resistance profiles. As determined from analyses of transfected cell lines, MRP1 is able to confer resistance to anthracyclines, vinca alkaloids, epipodophyllotoxins, camptothecins and methotrexate, but not to taxanes, which are an important component of the Pgp profile (Cole et al., 1994; Zaman et al., 1994; Breuninger et al., 1995; Chen et al., 1999a). Investigations of fibroblast cell lines derived from mrp1−/− mice have largely confirmed this resistance pattern (Allen et al., 2000; Johnson et al., 2001; Lin et al., 2002), but also showed that deficiency of the pump is associated with modest sensitization toward taxanes and mitoxantrone. However, there is currently little direct evidence indicating that the human protein has significant activity toward these two agents. Numerous reports document the expression of MRP1 in cancers that are treated with anthracyclines, camptothecins and etoposide, such as leukemia and breast, colorectal and germ cell, respectively, and in some cases correlations between clinical outcome and expression have been drawn (for examples of recent studies see Ohishi et al., 2002; Sauerbrey et al., 2002; van den Heuvel-Eibrink et al., 2002; Burger et al., 2003; Plasschaert et al., 2003; Zurita et al., 2003). It is reasonable to infer that MRP1 contributes to the inherent sensitivity of cancers in which it is expressed. However, there is currently no consensus with respect to its role in acquired resistance, or its prognostic significance.

In spite of the similarity in the resistance profiles of Pgp and MRP1, the substrate selectivities of the pumps differ markedly, in that whereas Pgp substrates are neutral or mildly positive lipophilic compounds, MRP1 is able to transport lipophilic anions (Figure 3a). Well-characterized MRP1 substrates include structurally diverse glutathione, glucuronate and sulfate conjugates, such as the cysteinyl leukotriene LTC4, a particularly high-affinity substrate and an important mediator of inflammatory responses, the estrogen glucuronide E217βG and sulfated bile acids (Leier et al., 1994; Jedlitschky et al., 1996; Loe et al., 1996a). Glutathione and glucuronate conjugates have been of particular interest in characterizations of MRP1 because they represent the products of phase II of cellular detoxification of hydrophobic xenobiotics, and efflux pumps involved in their cellular extrusion (phase III), which have been referred to as GS-X pumps in the case of glutathione conjugates, had been biochemically characterized in many cell types (Ishikawa, 1992). The ability of MRP1 to transport glutathione conjugates, in combination with its widespread expression in tissues, indicates that it is a ubiquitous GS-X pump (Kruh et al., 1995; Flens et al., 1996). As a cellular detoxifying factor, the subcellular localization of MRP1 in polarized epithelial cells is of interest. In contrast to Pgp, which extrudes xenobiotics into bile, intestine and blood for terminal elimination from the body, MRP1 is a basolateral transporter whose operation results in the movement of compounds away from luminal surfaces and into tissues that lie beneath the basement membrane (Figure 2) (Evers et al., 1996).

Figure 3

Involvement of glutathione in MRP1-mediated transport. (a) Transport of hydrophobic compounds that are enzymatically conjugated to glutathione. (b) Cotransport of etoposide with glutathione. (c) Transport of estrone 3-sulfate is stimulated by glutathione, but cotransport of glutathione has not been observed. (d) Transport of glutathione is stimulated by verapamil, but verapamil does not appear to be a transport substrate. (e) Transport of oxidized glutathione (GSSG)

The preference of MRP1 for lipophilic anions is reflected in the biochemical mechanism whereby the pump transports unmodified natural product anticancer agents, in that free glutathione is required for this process. A consequence of this is that MRP1-conferred resistance is subject to inhibition by buthionine sulfoximine (BSO), an agent that blocks the synthesis of glutathione by inhibiting γ-glutamylcysteine synthetase (γGCS) (Schneider et al., 1995; Versantvoort et al., 1995; Zaman et al., 1995). While this feature of the pump indicates that it lies at the intersection of thiol-based cellular resistance mechanisms and plasma membrane efflux systems, the precise mechanism by which glutathione participates in MRP1-mediated efflux of natural product drugs is currently unsettled. The predominant working model is that agents such as vinca alkaloids and anthracyclines are cotransported with glutathione (Figure 3b). This model is supported by studies showing that: (i) transport of vincristine and anthracyclines in membrane vesicle uptake assays is dependent upon glutathione (Loe et al., 1996b); (ii) etoposide stimulates glutathione extrusion in wild-type but not mrp1-deficient ES cells (Rappa et al., 1997); (iii) MRP1 can be photolabeled with glutathione analogs (Ciaccio et al., 1996); (iv) glutathione stimulates vanadate-induced trapping of 8-azido-ATP (Taguchi et al., 1997); and (v) glutathione derivatives lacking a free sulfhydryl group can support MRP1-mediated transport (Loe et al., 1998). Alternative possibilities involving allosteric interactions are also under consideration.

While it was initially thought that glutathione participated in the transport of hydrophobic compounds only, recent studies have disclosed a more complex picture. In addition to uncharged compounds, certain anionic conjugates such as estrone-3 sulfate and the glucuronide conjugate of a nitrosamine metabolite, NNAL-O-glucuronide, are also dependent upon glutathione, and transport of etoposide-glucuronide, which can be measured in the absence of glutathione, is enhanced by the tripeptide (Sakamoto et al., 1999; Leslie et al., 2001; Qian et al., 2001). Interestingly, transport of the former two compounds does not appear to be associated with cotransport of glutathione, and is therefore considered to be the result of a positive allosteric effect by glutathione (Figure 3c). Infrared spectroscopic measurements showing that glutathione binding induces conformational alterations in MRP1 are in accord with this interpretation (Manciu et al., 2003). Two other types of interactions involving glutathione have been described. Some compounds, such as the Pgp inhibitor verapamil, and certain bioflavonoids, are able to stimulate transport of glutathione by MRP1, but are not transport substrates themselves (Loe et al., 2000; Leslie et al., 2003a). Hence, these compounds exert an allosteric effect that increases the affinity of the pump for glutathione (Figure 3d). In addition, GSSG, the oxidation product of glutathione, is a good MRP1 substrate (Figure 3e) (Leier et al., 1996a). The latter finding is of significance because GSSG, which is normally present in low concentrations in the cell, is formed as a result of oxidative stress, and its cellular extrusion under these conditions is thought to represent a homeostatic mechanism that preserves normal redox balance. The involvement of MRP1 in this process is supported by experiments showing that MRP1 inhibitors diminish cellular extrusion of GSSG in rat astrocyte cells in which the pump is endogenously expressed (Hirrlinger et al., 2001).

Two types of anticancer agents are transported as glutathione complexes or conjugates by MRP1. MRP1 is able to confer resistance to heavy metal oxyanions such as arsenite and antimony (Cole et al., 1994; Rappa et al., 1997), and one report showed that the pump can be induced as a resistance factor in GLC4/Sb30, a cell line made resistant to antimony (Vernhet et al., 1999). This capability may be of clinical significance with regard to arsenic trioxide, an agent with great efficacy in the treatment of acute progranulocytic leukemia. A report, showing that GLC4/Sb30 and another MRP1-overexpressing cell line (GLC4/Adr) exhibit 8–10-fold levels of crossresistance to arsenic trioxide (Vernhet et al., 2001), suggests that the pump is able to confer resistance to this agent. It is presumed that MRP1 confers resistance to oxyanions such as arsenic trioxide by transporting these compounds as glutathione complexes, by analogy with the Leishmania MRP1 homologue PGPA and the yeast MRP1 homologue YCF1, which transport arsenite and cadmium, respectively, as glutathione complexes (Li et al., 1997; Legare et al., 2001). The ability of MRP1 to transport glutathione conjugates is also germane to chlorambucil, in that the pump is able to mediate the transport of monoglutathionyl-chlorambucil (Paumi et al., 2001). Interestingly, while most studies have found that MRP1 is unable to confer resistance to alkylating agents such as chlorambucil, the activity of the pump toward chlorambucil can be made manifest by coexpression with GSTα1-1, which catalyses the formation of monoglutathionyl-chlorambucil from the parent drug (Morrow et al., 1998). This finding may be of significance in that coordinate induction of MRP1 with other components of the thiol-based detoxification pathway has been observed (Kuo et al., 1996). Along the same lines, coexpression of MRP1 with γGCS, the pace-setter enzyme in glutathione biosynthesis, confers enhanced levels of heavy metal resistance (Lorico et al., 2002), and ectopic expression of MRP1, GSTπ and γ-GCS was reported to result in increased levels of resistance to etoposide and doxorubicin (O'Brien et al., 2000).

MRPs, methotrexate and reduced folates

Many aspects of the cellular pharmacology of methotrexate have been determined since its introduction as a chemotherapeutic agent more than 30 years ago (see review by Zhao and Goldman, in this issue). MTX, a monoglutamate, enters cells primarily via the reduced folate carrier, and is metabolized intracellularly by the addition of γ-linked glutamate residues in a reaction catalysed by FPGS. This reaction is crucial to the activity of this agent, because MTX polyglutamates are at least as potent as MTX in inhibiting DHFR, but in contrast to the parent compound, which is subject to energy-dependent efflux, are retained within the cell. Metabolic trapping of physiological folates is similarly a consequence of polyglutamylation. The energy-dependent system responsible for effluxing MTX, but not its intracellular metabolites, is therefore linked to the activity of this agent, but the molecular components of this system had not been identified until recently. It is now known that no less than four MRPs are components of this system. Analyses of the in vitro transport capabilities of MRPs 1–4, with respect to MTX versus MTX polyglutamates, and folates, indicate that these pumps have precisely the characteristics predicted by prior biochemical studies, in that: (i) they are high-capacity, low-affinity transporters of MTX (Km=2.2, 0.62 and 0.22 mM, for MRP1, MRP3 and MRP4, respectively); (ii) their capacity to efflux this agent is abrogated by the addition of even one glutamate residue; and (iii) they are similarly able to mediate the high-capacity, low-affinity transport of folates, such as folic acid and the reduced folate leucovorin (Km values for folic acid and leucovorin=2.0 and 1.7, and 0.17 and 0.64 mM, for MRP3 and MRP4, respectively) (Zeng et al., 2001; Chen et al., 2002). The involvement of MRPs in the cellular pharmacology of MTX is shown in Figure 4.

Figure 4

Schematic depicting the involvement of MRPs and ABCG2 in methotrexate efflux. Methotrexate (MTX) uptake into the cell is mediated by the reduced folate carrier (RFC1). Inside the cell, methotrexate is glutamylated by folyl-poly-γ-glutamate synthetase (FPGS) to yield MTX-Glu2 initially and eventually yield MTX-Glu3-7. Free methotrexate is extruded by MRP1, MRP2, MRP3 and MRP4, whereas MTX, MTX-Glu2 and MTX-Glu3 are subject to efflux by ABCG2 (adapted from Zeng et al., 2001)

These transport characteristics are reflected in the unusual MTX resistance phenotype conferred by MRPs. MRP1, MRP2 and MRP3, and to a lesser extent MRP4, are potent resistance factors when transfected cells are analysed in growth assays in which exposure to relatively high MTX concentrations is limited to the first few hours of the assay, but not in standard continuous drug exposure assays in which 100–1000-fold lower concentrations are employed (Hooijberg et al., 1999; Kool et al., 1999b; Lee et al., 2000). This phenotype is consequent to the ability of the pumps to reduce the accumulation of polyglutamates when drug exposure is brief, whereas under prolonged exposure conditions levels of polyglutamates sufficient to inhibit DHFR accumulate irrespective of the presence of overexpressed MRPs. The low affinities of the pumps may also contribute to this phenotype. At the low nM MTX concentrations employed in continuous exposure assays, the pumps may be out-competed by the higher affinity FPGS (Km40 μ M) (Sanghani and Moran, 2000), and once MTX is metabolized to polyglutamates the presence of ectopically expressed MRPs is of no consequence. MRPs are probably more effective at preventing metabolic trapping at the low μ M MTX concentrations employed in short-exposure assays, although these concentration are 10–100-fold lower than the Km values of the pumps.

The ability of MRPs to transport reduced folates suggested that expression of these pumps might influence cellular folate homeostasis, and this possibility has been substantiated by two recent reports. Ectopic expression of MRP1 and MRP3 was shown to reduce intracellular folate pools, and to render transfected cells susceptible to impaired growth when maintained in low folate medium (Hooijberg et al., 2003). Conversely, loss of MRP1 expression in a leukemia cell line (CEM-7A) established by gradual deprivation of folates was determined to represent an adaptive response that augments cellular folate pools (Assaraf et al., 2003). In combination these studies indicate that intracellular folate pools reflect a balance between influx via RFC1 and, at least in the case of CEM cells, efflux mediated in large part by MRP1.

At least one ABC transporter that is not an MRP family member is also a component of the MTX efflux system. Recently it was determined that expression of ABCG2 (BCRP, MXR), an efflux pump that is able to confer high levels of resistance to mitoxantrone and camptothecins, is also able to mediate MTX resistance (Volk et al., 2000, 2002). However, the transport characteristics of this pump with regard to folates and antifolates are distinct from MRPs. While ABCG2 is a high-capacity, low-affinity transporter of MTX (Km=1.3 mM), and its ability to transport this agent is attenuated by polyglutamylation, in contrast to MRPs, ABCG2 is able to transport MTX species having up to three glutamate residues (Chen et al., 2003a) (Figure 4). In addition, it appears that while ABCG2 is able to transport folic acid, it is unable to transport leucovorin.

Although certain drug-selected cell lines that have extremely high levels of ABCG2 expression, such as the mitoxantrone resistant cell line MCF7/MX, exhibit striking levels of crossresistance to MTX (i.e., 150-fold for MCF7/MX) in continuous drug exposure assays, ectopic expression of the pump in MCF-7 cells was reported to confer only 2-fold resistance to this agent in similar growth assays (Volk et al., 2002). Hence it appears that the ability of ABCG2 to transport lower polyglutamyl MTX species is not associated with enhanced potency as a resistance factor, at least as assessed in continuous drug exposure assays. Whether MTX resistance conferred by ABCG2 is increased in time-dependent drug exposure assays has not been determined. In considering the involvement of ABCG2 and MRPs in MTX efflux, it is important to bear in mind that while these pumps are able to confer MTX resistance in cell lines in which they are ectopically expressed, and inactivation of mrp1 renders fibroblasts sensitive to this agent (3-fold) (Lin et al., 2002), enhanced cellular extrusion has not been identified as a resistance factor in MTX-resistant cell lines or in patient samples (see review by Zhao and Goldman, in this issue).

In vivo functions of mrp1 as determined from investigations of gene-disrupted mice

mrp1+/− mice are healthy and fertile, indicating that the pump is dispensable for normal growth and development, at least under controlled laboratory conditions. Investigations of mrp1-deficient mice have confirmed several functions suggested by the in vitro studies described above. The capacity of mrp1 to function as an in vivo resistance factor for anticancer agents, is supported by the finding that mrp1-deficient mice are hypersensitive to etoposide (Lorico et al., 1997; Wijnholds et al., 1997). This phenotype is attributable to increased bone marrow toxicity in that mrp1−/− mice exhibit prolonged neutropenia (Lorico et al., 1997). In addition, mrp1-deficient mice challenged with etoposide were subject to increased tissue damage in the seminiferous tubules of the testes and oropharyngeal mucosal surfaces, where the protein is localized in Sertoli cells and the basal layer of epithelial cells, respectively. Analysis of mice that are deficient in both pgp1a/1b and mrp1 indicates that mrp1 is also able to reduce etoposide penetration into the cerebrospinal fluid by 10-fold (Wijnholds et al., 2000a). The involvement of mrp1 at this site is in accord with the localization of the pump in the basal membranes of choroid epithelial cells, and in vitro experiments implicating the pump as a barrier to basal → apical transepithelial drug permeation (Rao et al., 1999). Whether the effect of mrp1 on drug penetration into cerebrospinal fluid is significant in Pgp competent mice has not been determined. Interestingly, mdr1a−/−/1b−/−,mrp1−/− mice were found to be a striking 128-fold more sensitive to vincristine than wild-type mice, but only 3.5-fold more sensitive to etoposide, indicating that the magnitude of the combined defense provided by these pumps can be both synergistic and drug-specific (Johnson et al., 2001). These mice sustained marked intestinal and bone marrow toxicity. Greater sensitivity to vincristine than to etoposide was also apparent when fibroblasts derived from mdr1a−/−/1b−/−,mrp1−/− mice were analysed, but not nearly to the extent observed in vivo, indicating that tissue specificity is also a factor (Allen et al., 2000; Johnson et al., 2001; Lin et al., 2002). In combination, these studies indicate that mrp1 is a resistance factor for bone marrow and certain mucosal surfaces, that it constitutes a barrier in testes and choroid, and that normal tissues can be rendered markedly hypersensitive when deficiency of mrp1 is combined with inactivation of mdr1a/1b.

Investigations of mrp1−/− mice have confirmed the involvement of the pump in glutathione homeostasis and in inflammatory processes mediated by LTC4. The ability of mrp1 to influence intracellular glutathione levels in vivo, presumably by cotransporting glutathione with endogenous substrates, is indicated by measurements showing that levels in mrp1-deficient mice are elevated 25–90% in tissues in which the pump is known to be expressed, including breast, lung, heart, kidney, muscle, colon, testes, bone marrow cells, mononuclear leukocytes and erythrocytes, but are unchanged in organs in which expression of the pump is low, such as the liver and small intestine (Lorico et al., 1997). In vitro studies implicated mrp1 in the extrusion of LTC4 from mast cells (Figure 5) (Leier et al., 1996b; Bartosz et al., 1998), and this function was evident in vivo in that reduced cellular extrusion of this compound was found in bone marrow-derived mast cells prepared from mrp−/− mice, and the mice exhibited decreased ear edema in an immune assay in which LTC4-mediated inflammation is provoked by the topical application of arachidonic acid (Wijnholds et al., 1997). However, a seemingly paradoxical immunological phenomenon has been described in mrp1−/− mice challenged with Streptococcus pneumoniae-induced pneumonia (Schultz et al., 2001). In contrast to disruption of the 5-lipooxygenase gene (5-LO), which results in deficiency of both LTB4 and cysteinyl leukotrienes (Figure 5), and leads to enhanced lethality from bacterial pneumonia (Schultz et al., 2001), mrp1−/− mice are resistant to this infection. This unexpected phenotype was attributed to enhanced synthesis and extrusion in mrp1−/− mice of LTB4, a potent stimulant of phagocyte-mediated microbicidal activity, consequent to build up of intracellular LTA4 when efflux of LTC4 is impaired. This conclusion was supported by the findings that resistance to pneumonia in mrp1-deficient mice can be abrogated by treatment with specific LTB4 receptor antagonists, and that increased lethality in 5-LO−/− mice can be attenuated by provision of LTB4.

Figure 5

Biosynthesis of leukotrienes and extrusion of LTC4 by MRP1. Leukotriene A4 (LTA4) is synthesized by the successive actions of cytoplasmic phospholipase A2 (cPLA2) and 5′-lipoxygenase. Conjugation of LTA4 with glutathione yields the cysteinyl leukotriene LTC4, which is effluxed by MRP1. Following extrusion, LTC4 is converted to LTD4. LTC4 and LTD4 mediate inflammation by binding to leukotriene receptors on target cells. The noncysteinyl leukotriene LTB4, which is also a potent mediator of inflammation, is also synthesized from LTA4

While the involvement of MRP1 in inflammatory responses mediated by cysteinyl leukotrienes had been anticipated, investigation of mrp1−/− mice has disclosed an unexpected role for these compounds in dendritic cell function. These antigen-bearing cells participate in primary immune responses by acquiring antigens in the periphery and tracking via afferent lymphatic vessels to draining lymph nodes, where they localize in T-cell-rich regions and stimulate the expansion of naïve T cells. In an experimental model of contact sensitivity involving the application of FITC to the skin, mrp1−/− mice retained FITC-positive dendritic cells in the skin, and accumulated significantly fewer positive dendritic cells in lymph nodes (Robbiani et al., 2000). This finding, in combination with the determination that provision of LTC4 could overcome this defect and that LTC4 is required for dendritic cells to exhibit an optimal response to chemotactic ligands in vitro, suggests a schema in which triggering of leukotriene receptors on dendritic cells stimulates migration of these cells to afferent lymphatic vessels (Randolph, 2001). Whether the cells responsible for extrusion of LTC4 are dendritic cells or other cell types, such as mast cells, is currently unclear.

Structural studies

The topology of the N-terminal extension of MRP1 (MSD0 and L0), a striking structural feature of the pump, has been experimentally determined (Bakos et al., 1996; Hipfner et al., 1997; Kast and Gros, 1997), whereas the configuration of transmembrane helices in MSD1 and MSD2 is based upon computer algorithms. The MSD0 domain is grossly dispensable for function, as an N-terminal truncated mutant that lacks this domain is functional with respect to membrane vesicle transport activity, susceptibility to vanadate-induced nucleotide trapping, assuming basolateral localization in polarized cells and mediating cellular efflux of daunorubicin and glutathione conjugates (Bakos et al., 1998). However, extending the N-terminal truncation to include the L0 domain abrogates the activity of the pump, indicating that this domain is essential for function. The indispensability of the L0 domain is in accord with its presence in all MRP family members (Figure 1). Notwithstanding evidence indicating that N-terminal deletion of the entire MSD0 does not alter several properties of the pump, it is anticipated that this domain has functions that have yet to be uncovered. Studies showing that MRP1 activity can be affected by point mutations in the extracellular portion of the N-terminus and in MSD0, and by N-terminal deletions that extend to and include the first transmembrane domain hint at this possibility, although it remains to be determined whether these perturbations influence actual drug binding and transport as opposed to preventing the assumption of correct topology in the plasma membrane (Gao et al., 1998; Yang et al., 2002; Leslie et al., 2003b).

Drug binding sites on MRP1 have been explored by the use of photoaffinity labeling, and these studies have implicated sites in MSD1 and MSD2. Experiments employing the quinoline-based drug IACI and the rhodamine analog IAARh123 identified TM10–11 in MSD1 and TM16–17 in MSD2 as drug binding determinants (Daoud et al., 2000a, 2000b, 2001). Sites that overlap these two regions are also crosslinked by the high-affinity substrate LTC4, and LY474776, a glutathione-dependent inhibitor of the pump (Mao et al., 2002; Qian et al., 2002). LTC4 binding sites were mapped to a region encompassing MSD1-NBD1 and to TM14–17, and LY474776 binding was mapped to TM16–17. In addition, agosterol A, a polyhydroxylated sterol acetate that is a potent glutathione-dependent inhibitor, was also reported to crosslink TM17 (Ren et al., 2001, 2002). The results of site-directed mutagenesis studies support the involvement of TM14 and TM17 in MRP1 activity, and have also implicated residues in TM6 (Ito et al., 2001b; Zhang et al., 2001a, 2001b, 2002; Haimeur et al., 2002; Ren et al., 2002).

Recent reports have also begun to probe the structural determinants of the interaction of glutathione with MRP1. Glutathione-dependent binding of agosterol A to the COOH half of the protein required the presence of the L0 domain, implicating the latter region in glutathione binding (Ren et al., 2001). MRP1 can be crosslinked with glutathione photoaffinity reagents (Ciaccio et al., 1996), and studies using these probes have provided more direct information on sites of glutathione interaction. Binding of IAAGSH, a [I125]labeled azido-derivative of glutathione, was mapped to the L0 and L1 domains, in addition to regions TM10–11 and TM16–17, which were previously implicated as drug binding sites (Karwatsky et al., 2003). Another study employing a different photoactive glutathione derivative, azidophenacyl-GSH, described crosslinking to both the N- and C-terminal halves of the protein (Qian et al., 2002). While crosslinking to the L0 domain was not observed, this region was implicated in that it was required for labeling, similar to the situation with GSH-dependent binding of agosterol A.

In combination, these initial studies suggest that hydrophobic drugs, which may be present in the lipid bilayer, interact with sites in the transmembrane domains of MSD1 and MSD2, and that cytoplasmic glutathione may interact with at least some MRP1 sequences that are intracellular. Images of MRP1 obtained from single particle image analysis and electron microscopy of two-dimensional crystals suggest that the reconstituted protein is dimeric (Rosenberg et al., 2001). However, the resolution of the currently available images is too low to permit inferences concerning how the pump binds and transports substrates.


Canalicular efflux pump for amphipathic anions and resistance factor for anticancer agents

The substrate selectivity of MRP2 is similar to that of MRP1 with respect to glutathione and glucuronate conjugates, but recent reports indicate that the transport characteristics of the pumps differ in detail (Cui et al., 1999; Kawabe et al., 1999). MRP2 is a lower affinity transporter for conjugates, and MRP2-mediated transport of compounds such as E217βG, an established physiological substrate of the pump (Morikawa et al., 2000), is subject to positive allosteric regulation by bile acids and certain other amphipathic anions (Bodo et al., 2003; Zelcer et al., 2003a). In spite of the similarity in substrate range, the functions of MRP2 in the body are distinct from those of MRP1 as a result of differences in expression pattern and subcellular polarity. In contrast to MRP1, MRP2 assumes apical localization in polarized cells (Figure 2), and it is mainly expressed in liver canaliculi, with lower levels in renal proximal tubules, gut enterocytes, syncytiotrophoblast cells of the placenta and possibly brain capillaries (Kartenbeck et al., 1996; Schaub et al., 1997; Miller et al., 2000; Mottino et al., 2000; St-Pierre et al., 2000). Therefore, it is functionally similar to Pgp in its involvement in the terminal elimination of compounds and its role as a barrier in gut and placenta.

Prior to the molecular identification of MRP2, several of its cardinal biochemical and physiological functions had been deduced from investigations of humans and rats that have since been determined to be genetically deficient in the pump (Kartenbeck et al., 1996; Paulusma et al., 1996). In older studies, the protein that is now known as MRP2 was often referred to as the canalicular multispecific organic anion transporter (cMOAT), a designation that aptly describes its ability to extrude a range of lipophilic anions into the bile. The human condition, Dubin–Johnson syndrome, is a largely asymptomatic disorder whose principal manifestation is jaundice (Dubin and Johnson, 1954). This abnormality reflects the role of MRP2 in the biliary excretion of bilirubin glucuronide, a conjugate that results from the action of hepatic UDP-glucuronosyl transferases on the end product of heme degradation (Jedlitschky et al., 1997). Familial deficiency of MRP2 is accompanied by defects in the hepatobiliary excretion of a variety of other anionic compounds, such as sulfobromophthalein (BSP), a diagnostic dye that is extruded as a glutathione conjugate by MRP2. Rodent models of Dubin–Johnson syndrome (EHBR and GY/TR rat strains) are also well characterized, and have been particularly valuable in investigations of the involvement of mrp2 in biliary extrusion of pharmaceutical agents (Jansen et al., 1985; Kitamura et al., 1990).

Investigations of cell lines in which either the recombinant protein or antisense constructs are expressed indicate that the drug resistance profile of MRP2 is similar to that of MRP1 with respect to anthracyclines, vinca alkaloids, epipodophyllotoxins and camptothecins (Koike et al., 1997; Cui et al., 1999; Kawabe et al., 1999). However, MRP2 appears to have somewhat reduced potency toward these agents, and it has not been reported to be overexpressed in cell lines made resistant to natural product agents, a circumstance that is well described for MRP1, Pgp and ABCG2, and speaks to the importance of the latter pumps as deployable resistance factors in cellular models. Another difference between MRP1 and MRP2 is that the latter pump is able to confer resistance to cisplatin, an agent that is known to form toxic glutathione conjugates in the cell (Ishikawa and Ali-Osman, 1993), and overexpression of MRP2 has been described in at least one cisplatin-resistant cell line (Taniguchi et al., 1996). Similar to the situation with MRP1, glutathione plays a role in MRP2-mediated transport of hydrophobic anticancer agents, as indicated by the ability of vinblastine to stimulate glutathione efflux from transfected MDCK cells, and the ability of (rabbit) MRP2 to mediate transport of this drug in membrane vesicle assays only in the presence of glutathione (Van Aubel et al., 1999; Evers et al., 2000). That glutathione is an MRP2 substrate in vivo is indicated by its absence in the bile of mrp2-deficient rats (Elferink et al., 1989).

Although its significance as an in vivo resistance factor remains to be determined, expression of MRP2 has been reported for several types of human cancers that are relevant to the pump's resistance profile, including colorectal (camptothecins), breast and leukemia (anthracyclines), ovary (cisplatin) and others (for example, Hinoshita et al., 2000; Nies et al., 2001; Ohishi et al., 2002; Burger et al., 2003; Steinbach et al., 2003, and references therein). MRP2 plays a role in the hepatobiliary excretion of numerous pharmaceuticals (Suzuki and Sugiyama, 2002). In addition, the functionality of the pump in the gut and possibly brain capillaries has been inferred from studies showing that mrp2-deficient rats have decreased extrusion of heterocyclic amines into the intestine and increased levels of phenytoin in the brain (Dietrich et al., 2001; Potschka et al., 2003). With respect to cancer chemotherapeutics, the involvement of MRP2 in hepatobiliary elimination is indicated by studies showing that mrp2-deficient rat strains have reduced biliary extrusion of methotrexate and CPT-11 (Chu et al., 1997; Masuda et al., 1997), and that cyclosporine A, a Pgp modulator that also inhibits MRP2 (Chen et al., 1999b), reduces biliary clearance of the latter agent (Gupta et al., 1996).

Structural studies

Photoaffinity labeling studies have not as yet been reported for MRP2. However, inferences about substrate binding sites have been made on the basis of site-directed mutagenesis. These studies indicate the involvement of TM6, TM9, TM16 and TM17 in the human protein and TM11, TM14 and TM16 for the rat (Ryu et al., 2000; Ito et al., 2001a, 2001c, 2001d). Many of the structure–function studies reported for MRP2 to date have involved investigations of the determinants that mediate apical sorting. Analysis of chimeric MRP1/MRP2 proteins showed that a 480-amino-acid NH2-terminal segment of MRP1 was sufficient for localization of the protein to lateral membranes of MDCK cells, suggesting the presence of an apical targeting sequence in the analogous region of MRP2 (Konno et al., 2003). In accord with the presence of an N-terminal apical routing determinant, an N-terminal MRP2 deletion mutant lacking the MSD0 and L0 domains failed to target to the apical membranes of MDCK cells, and instead localized intracellularly (Fernandes et al., 1990). However, two other reports point to the possibility of C-terminal determinants. Apical routing of CFTR involves an interaction with a PDZ scaffolding protein that is mediated by a PDZ interaction motif at the C-terminus of CFTR (Moyer et al., 1999). Recently, MRP2 was also determined to have a functional PDZ interaction motif (‘TKF’) at its extreme C-terminus, as indicated by yeast two-hybrid and in vitro overlay assays (Kocher et al., 1999; Hegedus et al., 2003), and deletion or mutation of this sequence in MRP2-GFP was reported to confer basolateral localization in transfected MDCK cells (Harris et al., 2001). However, another study using a similar design found that truncation of 15 C-terminal amino acids from a GFP-MRP2 protein was necessary for loss of apical polarization in HepG2 cells, whereas truncation of the 3-amino-acid PDZ motif alone had no effect (Nies et al., 2002a). In addition, in the latter study an MRP1/MRP2 chimeric protein with a junction at amino acid 792 was reported to assume apical localization in transfected MDCK cells, a finding that conflicts with the study of Konno et al., in which a similar MRP1/MRP2 protein (junction at amino acid 846) was found to have lateral localization in transfected LLC-PK1 cells. The complex picture that has emerged from these initial and in some cases conflicting reports may be consequent to the involvement of factors that may be cell-specific and that may not only govern routing per se, but also determine the stability of a membrane localized protein, and the ability of an abnormal protein to escape the intracellular quality control system.


Resistance to etoposide

Among MRP family members, MRP3 has the highest degree of structural resemblance to MRP1 (58%), and as would be expected, its substrate selectivity overlaps with that of MRP1 and MRP2 with respect to the transport of glutathione and glucuronate conjugates (Hirohashi et al., 1999; Zeng et al., 1999). However, the affinity of MRP3 for conjugates is significantly lower than those of MRP1, and its drug resistance capabilities are not as extensive as either MRP1 or MRP2. With respect to natural product agents, analyses of cell lines in which MRP3 has been ectopically expressed indicate that the pump is probably only able to confer low levels of resistance to etoposide and teniposide (Kool et al., 1999b; Zeng et al., 1999; Zelcer et al., 2001). Another difference is that MRP3, in contrast to MRP1 and MRP2, does not appear to require glutathione for mediating the transport of natural product agents, in that BSO does not attenuate its potency as a resistance factor (Zelcer et al., 2001). The finding that MRP3-transfected cells do not have depressed levels of glutathione is also in accord with the possibility that glutathione is not a substrate of the pump (Kool et al., 1999b). These features suggest that the limited resistance properties of MRP3 may relate to its lower affinity for amphipathic anions in general and glutathione in particular.

Back-up system for amphipathic anions in cholestatic conditions

MRP3 is able to transport monoanionic bile acids such as glycocholate and taurocholate, which constitute a significant component of bile acids in humans and rodents (Hirohashi et al., 2000; Zeng et al., 2000). This capability has drawn considerable attention because MRP3, which is usually expressed at low levels at the basolateral surfaces of bile duct cells and hepatocytes, is dramatically induced during cholestatic conditions (see Hirohashi et al., 1998; Donner and Keppler, 2001; Soroka et al., 2001, and references therein). In combination, these features suggest that when the usual canalicular route of excretion is blocked, MRP3 may function to detoxify hepatocytes of bile acids and other conjugates by mediating the extrusion of these compounds into sinusoidal blood (Figure 6). (A recent study showed that MRP2 is also able to transport glycocholate, but the absence of reported bile acid abnormalities in MRP2-deficient humans and rats suggests that the bile salt export pump (BSEP, SPGP) is predominately responsible for this apical activity under normal conditions (Gerloff et al., 1998; Bodo et al., 2003).) The determination that MRP3 can be induced by bile acids, at least in the context of an enterocyte cell line (Caco2), and the identification of MRP3 promoter elements that mediate bile acid induction suggest that the increased levels of bile acids consequent to cholestatic conditions may directly contribute to upregulation of the pump in hepatocytes (Inokuchi et al., 2001). It has also been speculated that MRP3 may be involved in the enterohepatic circulation of bile acids, in that the pump is localized to the basolateral surfaces of enterocytes (Rost et al., 2002), where it may mediate the transport into blood of bile acids carried into the gut by the bile and taken up across the apical surfaces of enterocytes by the ileal Na-dependent bile salt transporter. However, the observation that mrp3−/− mice are healthy (M Belinsky and G Kruh, unpublished data) suggests that MRP3 is not likely to play an important role in this essential process. In addition to gut and liver, MRP3 is expressed in a variety of other tissues, including pancreas, kidney, adrenal and gallbladder (Belinsky et al., 1998; Kiuchi et al., 1998; Uchiumi et al., 1998; Scheffer et al., 2002b). The functions of the pump in these and other tissues may be revealed by investigations of mrp3-deficient mice.

Figure 6

Schematic depicting the proposed function of MRP3 during cholestasis. Under normal conditions, bile acids such as glycocholic acid and conjugates such as bilirubin glucuronide are extruded across the (apical) canalicular membrane of hepatocytes by the bile salt export pump (BSEP, SPGP) and MRP2, respectively. MRP3, which is located in basolateral membranes of hepatocytes (and cholangiocytes), is markedly induced in cholestatic conditions. The ability of MRP3 to mediate the transport of both monoanionic bile acids and conjugates suggests that it may function as a basolateral back-up system to extrude these compounds into sinusoidal blood when the canalicular route is blocked

MRP4, MRP5 and MRP8

Resistance factors for nucleotide analogs

The absence of a third (N-terminal) membrane spanning domain (Figure 1) suggested that MRP4 and MRP5 might have distinct properties (Belinsky et al., 1998), and this has proved to be the case with respect to both substrate selectivity and drug resistance capabilities. MRP4 and MRP5 are organic anion transporters, as indicated by the capacity of the former pump to transport prototypical MRP1 substrates such as E217βG, methotrexate and reduced folates, and the ability of the latter pump to efflux anionic fluorochromes (McAleer et al., 1999; Lee et al., 2000; Chen et al., 2001, 2002). In addition, intracellular glutathione levels are depressed in MRP4- and MRP5-transfected cells, indicating that this compound is also a substrate of the pumps (Wijnholds et al., 2000b; Lai and Tan, 2002). However, in contrast to the larger members of the family, MRP4 and MRP5 are able to mediate the transport of cAMP and cGMP (Jedlitschky et al., 2000; Chen et al., 2001; van Aubel et al., 2002). The substrate selectivity of MRP8 has not been described in detail in in vitro transport assays, but its ability to transport cyclic nucleotides is indicated by enhanced cellular extrusion in transfected cells (Guo et al., 2003). This capability appears to underlie the distinct drug resistance profiles of these pumps. MRP4, MRP5 and MRP8 do not confer resistance to natural product agents, but instead have the ability to confer resistance to certain nucleotide analogs. MRP4 was initially implicated in this process by the determination in the Fridland laboratory that MRP4 was overexpressed in a PMEA-resistant cell line that exhibits increased efflux of this anionic nucleotide analog (Figure 7a) (Robbins et al., 1995; Schuetz et al., 1999). Analysis of transfected cell lines further revealed that MRP4 and MRP5 are not only able to confer resistance to this acyclic nucleotide analog employed in the treatment of hepatitis B, but are also resistance factors for anticancer agents such as 6-mercaptopurine and 6-thioguanine, and in the case of MRP4, methotrexate and the antiviral ganciclovir (Lee et al., 2000; Wijnholds et al., 2000b; Chen et al., 2001; Adachi et al., 2002). MRP8 is also able to confer resistance to PMEA, as well as to fluoropyrimidines and the anti-AIDS agent ddC (Guo et al., 2003).

Figure 7

Efflux of purine and pyrimidine nucleotide analogs by MRP4, MRP5 and MRP8. (a) PMEA, an amphipathic anion, is directly effluxed by MRP4, MRP5 and MRP8. PMEA is activated to PMEApp, which inhibits hepatitis B reverse transcriptase (HBV RT). (b, c) 6-Mercaptopurine (6-MP) and 5′-fluorouracil (5-FU) are metabolized to their intracellular cytotoxic metabolites thiopurine nucleotide monophosphates (tNMPs) and 5-fluoro-2′-deoxyuridine monophosphate (5-FdUMP), respectively. tNMPs and 5-FdUMP are subject to efflux by MRP4, MRP5 and MRP8, as shown. The range of nucleotide analogs for which these pumps are able to confer resistance has not been fully determined

As would be expected for amphipathic anion pumps, these transporters do not efflux unmodified thiopurines and fluorpyrimidines, which are uncharged purine and pyrimidine base analogs, respectively. Rather, they transport the active nucleotide metabolites of these agents (Figure 7b,c). It has been inferred from the analysis of metabolites effluxed from MRP4- and MRP5-transfected cells that thiopurine nucleotides are the relevant substrates of these two pumps, and membrane vesicle transport studies indicate that MRP8 is able to transport 5-fluoro-2′-deoxyuridine monophosphate, but neither the parent compound nor its ribosylated intermediate metabolite (Wielinga et al., 2002; Guo et al., 2003). Established factors involved in resistance to thiopurines, agents whose primary use in cancer treatment is in the maintenance component of the treatment of childhood acute lymphoblastic leukemia, include polymorphic variations which reduce the activity of thiopurine methyltransferase, an enzyme that inactivates the parent drugs, and depressed expression levels of enzymes involved in anabolic activation of these agents (see review by Krynetski and Evans, in this issue). In the case of fluoropyrimidines, alterations in the levels of the target enzyme thymidylate synthase, and of activating enzymes, are established resistance factors. It remains to be determined whether these pumps are actually induced consequent to thiopurine or fluoropyrimidine exposure in treated patients, a circumstance that is currently not supported by studies of resistant cellular models. Similarly, while these pumps are potential cellular resistance factors for antivirals such as PMEA and ddC, further studies are needed to determine if the pumps are induced in patients treated with these agents.

Cyclic nucleotide efflux pumps

Energy-dependent cellular extrusion of cyclic nucleotides from eucaryotic cells is a well-established phenomenon (Bankir et al., 2002). However, efflux pumps have not been considered to be the major determinants of intracellular cyclic nucleotide levels because the precise time-sensitive signaling that is associated with these second messengers is thought to require immediate attenuation by the rapid, high-capacity system constituted by cellular phosphodiesterases (Soderling and Beavo, 2000). In contrast to the enzymatic breakdown of cAMP and cGMP, cellular extrusion is thought to be a relatively slow, low-capacity process. The determination that MRP5 and, more recently, MRP4 and MRP8 are able to transport these second messengers, revealed at least three of the membrane proteins that accomplish this process, and provided the molecular tools for examining the extent of the involvement of efflux pumps in cyclic nucleotide homeostasis (Figure 8) (Jedlitschky et al., 2000; Chen et al., 2001; Guo et al., 2003). Initial reports on the impact of ectopic expression of MRP4, MRP5 and MRP8 on cyclic nucleotide levels tend to support the view that they are not major determinants of intracellular cyclic nucleotide levels, in that only modest decreases in the levels of cAMP and cGMP were found (Lai and Tan, 2002; Guo et al., 2003; Wielinga et al., 2003). These studies, in combination with reports in which low micromolar affinities were measured in membrane vesicle transport assays for MRP4- and MRP5-mediated transport of cGMP, and for MRP4 in the case of cAMP, suggest that their effects on cyclic nucleotide levels are limited by the highly efficient phosphodiesterase system, as opposed to their inability to function at physiological concentrations (Jedlitschky et al., 2000; Chen et al., 2001).

Figure 8

Schematic diagram depicting cellular extrusion of cyclic nucleotides by MRP4, MRP5 and MRP8. cAMP and cGMP are synthesized when adenylyl (AC) and guanylyl cyclases (GC) are triggered by peptide ligands (L) and nitric oxide (NO), and degraded by the action of phosphodiesterases (PDEs). Cyclic nucleotides are also subject to cellular extrusion by a membrane system that includes MRP4, MRP5 and MRP8. However, the extent to which this process influences cyclic nucleotide homeostasis, and the physiological processes involving extruded cyclic nucleotides remain to be defined (adapted from Chen et al., 2001)

Cellular extrusion has also been implicated in the provision of extracellular cyclic nucleotides, which have been proposed to function as primary messengers (Bankir et al., 2002). In this respect, MRP4 is a good candidate for mediating the extrusion of cAMP in urine, based upon the apical localization of the rat protein in renal tubular cells (van Aubel et al., 2002). Investigation of the effects of these pumps has just begun, and further studies will be required to define the extent of their involvement in cyclic nucleotide homeostasis and in physiological processes attributed to extruded cAMP. In addition, glutathione levels are depressed in MRP4- and MRP5-transfected cell lines, indicating that this compound is a substrate of the pumps, but the involvement of glutathione in efflux mediated by the two pumps has not been explored (Wijnholds et al., 2000b; Lai and Tan, 2002).

Endogenous substrates other than cyclic nucleotides have been identified for MRP4, and it is possible that the pump may be involved in functions associated with these compounds. Aside from reduced folates, MRP4 is also able to transport glucuronide and sulfate conjugates of steroids, such as E217βG and DHEAS (Chen et al., 2001; Zelcer et al., 2003b). Transport of steroid conjugates may be of particular significance with respect to the prostate, or possibly prostate-derived tumors, because these tissues are hormonally responsive and express high levels of MRP4 transcript (Lee et al., 1998). In addition, MRP4 protein has been localized to the glandular epithelial cells in prostate (Lee et al., 2000). Aside from the original description of MRP5 as a cyclic nucleotide efflux pump, other endogenous substrates of the pump have not been conclusively identified. To understand the potential physiological functions of these pumps, more information about their tissue-specific expression patterns is needed. MRP4 has only been localized in prostate and kidney, as mentioned above, and MRP5 has only been localized to urogenital tissue (Nies et al., 2002b). The latter finding may be of pharmacological interest in view of the role of cGMP in erectile function, and membrane vesicle experiments showing that cGMP transport by MRP5 is susceptible to inhibition by phosphodiesterase inhibitors such as sildenafil, although this class of agents did not appear to be potent inhibitors of cyclic nucleotide extrusion by MRP5-transfected cells (Jedlitschky et al., 2000; Wielinga et al., 2003). The tissue distribution of MRP8 protein has yet to be described, but like MRP4 and MRP5, its transcript can be detected in many tissues (Bera et al., 2001; Tammur et al., 2001; Yabuuchi et al., 2001).


Pseudoxanthoma elasticum

A surprising discovery pertaining to MRP6 is that its genetic deficiency is the basis of pseudoxanthoma elasticum (PXE), a rare autosomally inherited connective disease whose predominant pathological findings are dystrophic elastin fibers in the skin, retina and large blood vessels, with corresponding clinical manifestations of baggy skin, loss of vision and calcification of large blood vessels (Bergen et al., 2000; Le Saux et al., 2000; Ringpfeil et al., 2000). The involvement of MRP6 in PXE was puzzling because RNA blot and RNase protection assays described in the initial characterizations of the human cDNA indicated that transcript expression was limited to liver and kidney with very low or undetectable levels in other tissues (Belinsky and Kruh, 1999; Kool et al., 1999a). A similar pattern of transcript expression was reported for the rat MRP6 homologue except that its expression in kidney and gut appeared to be lower and higher, respectively, that the case in humans (Hirohashi et al., 1998; Madon et al., 2000). In addition, an immunohistochemical analysis in which a monoclonal MRP6 antibody was employed showed that the human protein was abundant in liver and kidney – where it was localized to basolateral surfaces of hepatocytes and proximal tubules, but not expressed in many other tissues including those affected in PXE patients (ie, skin and retina) (Scheffer et al., 2002a). These findings raised the possibility that PXE might result from the absence of a substance that is normally extruded from liver or kidney into blood, and which is involved in connective tissue homeostasis (Uitto et al., 2001). However, in contrast to these studies, human MRP6 transcript was detected by RT/PCR in skin, blood vessels and retina, and recently reported in situ hybridization and immunohistochemical experiments on mouse mrp6 showed transcript and protein expression not only in liver and kidney, but also in many other tissues, including skin, retina and aorta (Bergen et al., 2000; Beck et al., 2003). As a result of these apparently conflicting reports on the human and rodent proteins, the tissue-specific site of MRP6 action that is relevant to PXE is currently unsettled.

In vitro properties

Reports on the functional properties of MRP6 indicate that the pump is able to transport lipophilic anions. The rat protein was reported to be competent in the transport of the anionic cyclopentapeptide BQ123 (Madon et al., 2000), and more recently the human protein was shown to transport glutathione conjugates such as LTC4 and N-ethylmaleimide-glutathione, in addition to BQ123, but not glucuronate conjugates such as E217βG (Belinsky et al., 2002; Ilias et al., 2002). While these studies have revealed that MRP6 is an amphipathic anion transporter, and that mutations reported in PXE patients compromise the transport activity of the pump, it remains to be determined whether transport of glutathione conjugates, or possibly glutathione, is relevant to the pathogenesis of PXE. Investigation of potential MRP6 substrates and analysis of MRP6 protein expression in affected tissues is ongoing in many laboratories, and should help to clarify these issues.

Analysis of MRP6-transfected CHO cells indicated that MRP6 is able to function as a drug pump (Belinsky et al., 2002). This initial study showed that MRP6 is able to confer low levels of resistance to etoposide and teniposide, but not to podophyllotoxin. In addition, low levels of resistance were detected for anthracyclines and cisplatin.

MRP7 and MRP9

Of the two recently described MRP family members, functional studies have only been reported for MRP7. In vitro experiments showed that this pump has the facility for mediating the transport of E217βG, and to a lesser extent LTC4, but not other MRP substrates such as cyclic nucleotides, methotrexate or bile acids (Chen et al., 2003b). Although RT/PCR analysis indicated that MRP7 is expressed in many tissues, transcript was difficult to detect by Northern blot analysis, suggesting that MRP7 expression may be low in most tissues (Hopper et al., 2001). The highest levels of murine MRP7 transcript were found in rat heart, liver, skeletal muscle and kidney (Kao et al., 2002). Functional studies on MRP9 have yet to be reported, but its structural resemblance to MRP4, MRP5 and MRP8 (Figure 1) raises the possibility that it may share some of the properties of the latter pumps. Two reports indicate that MRP9 transcript expression is relatively restricted, whereas a third study found more generalized expression (Tammur et al., 2001; Yabuuchi et al., 2001; Bera et al., 2002). MRP9 appears to be subject to an usual degree of alternative splicing, and this may account for differences in the results of RT/PCR analyses of expression. In addition, MRP9 transcript was also reported for nine of 12 breast cancer samples (Bera et al., 2002).


At least some of the functional properties of many members of the MRP family have now been determined. Physiological roles for MRP1 in protecting certain tissues from the effects of chemotherapeutic agents, and in inflammation and dendritic cell function, have been identified, and MRP2 is involved in the hepatobiliary elimination of bilirubin glueuronide and many pharmaceutical agents. Both of these pumps are potent resistance factors for natural product chemotherapeutic agents. MRP3 is able to confer resistance to epipodophyllotoxins and may have a role in the disposition of bile acids in pathological conditions. MRP1, MRP2, MRP3 and MRP4 are resistance factors for methotrexate, and MRP1 may be involved in folate homeostasis. MRP4, MRP5 and MRP8 are cyclic nucleotide efflux pumps that can been deployed for the purpose of conferring resistance to nucleotide analogs.

With respect to the treatment of cancer, many important questions about these pumps are unresolved. Within the panoply of resistance factors that are at the disposal of cancer cells (see other reviews, in this issue), the relative importance of pumps such as MRPs has not been determined. This is an issue that will probably be best evaluated when potent inhibitors of MRP1 are developed and applied in clinical trials. Given the potential for the closely related MRP2 to affect hepatobiliary elimination of the drugs that are part of the resistance profile of MRP1, the specificity of MRP1 inhibitors may also be an issue. In addition, it is now clear that there is considerable functional redundancy between certain MRPs, ABCG2 and Pgp, and this will have to be taken into account in the application of potential inhibitors (discussed in Kruh et al., 2001). Redundancy also represents a challenge in clinical studies aimed at making correlations between drug sensitivity and expression of specific pumps, which is exactly the type of information that is critical to determining which cancers and which treatments to target in clinical studies of inhibitors. Nevertheless, we are approaching a time when a more complete picture of membrane systems involved in the extrusion of chemotherapeutics agents is achieved, and this will no doubt help to focus clinical questions involving drug efflux pumps.


  1. Adachi M, Sampath J, Lan LB, Sun D, Hargrove P, Flatley R, Tatum A, Edwards MZ, Wezeman M, Matherly L, Drake R and Schuetz J . (2002). J. Biol. Chem., 277, 38998–39004.

  2. Allen JD, Brinkhuis RF, van Deemter L, Wijnholds J and Schinkel AH . (2000). Cancer Res., 60, 5761–5766.

  3. Assaraf YG, Rothem L, Hooijberg JH, Stark M, Ifergan I, Kathmann I, Dijkmans BA, Peters GJ and Jansen G . (2003). J. Biol. Chem., 278, 6680–6686.

  4. Bakos E, Evers R, Szakacs G, Tusnady GE, Welker E, Szabo K, de Haas M, van Deemter L, Borst P, Varadi A and Sarkadi B . (1998). J. Biol. Chem., 273, 32167–32175.

  5. Bakos E, Hegedus T, Hollo Z, Welker E, Tusnady GE, Zaman GJ, Flens MJ, Varadi A and Sarkadi B . (1996). J. Biol. Chem., 271, 12322–12326.

  6. Bankir L, Ahloulay M, Devreotes PN and Parent CA . (2002). Am. J. Physiol. Renal Physiol., 282, F376–92.

  7. Bartosz G, Konig J, Keppler D and Hagmann W . (1998). Biol. Chem., 379, 1121–1126.

  8. Belinsky MG, Bain LJ, Balsara BB, Testa JR and Kruh GD . (1998). J. Natl. Cancer Inst., 90, 1735–1741.

  9. Belinsky MG, Chen ZS, Shchaveleva I, Zeng H and Kruh GD . (2002). Cancer Res., 62, 6172–6177.

  10. Belinsky MG and Kruh GD . (1999). Br. J. Cancer, 80, 1342–1349.

  11. Bera TK, Iavarone C, Kumar V, Lee S, Lee B and Pastan I . (2002). Proc. Natl. Acad. Sci. USA, 99, 6997–7002.

  12. Bera TK, Lee S, Salvatore G, Lee B and Pastan IH . (2001). Mol. Med., 7, 509–516.

  13. Bergen AA, Plomp AS, Schuurman EJ, Terry S, Breuning M, Dauwerse H, Swart J, Kool M, van Soest S, Baas F, ten Brink JB and de Jong PT . (2000). Nat. Genet., 25, 228–231.

  14. Bodo A, Bakos E, Szeri F, Varadi A and Sarkadi B . (2003). J. Biol. Chem., 278, 23529–23537.

  15. Breuninger LM, Paul S, Gaughan K, Miki T, Chan A, Aaronson SA and Kruh GD . (1995). Cancer Res., 55, 5342–5347.

  16. Burger H, Foekens JA, Look MP, Meijer-van Gelder ME, Klijn JG, Wiemer EA, Stoter G and Nooter K . (2003). Clin. Cancer Res., 9, 827–836.

  17. Chen Z-S, Furukawa T, Sumizawa T, Ono K, Ueda K, Seto K and Akiyama SI . (1999a). Mol. Pharmacol., 55, 921–928.

  18. Chen Z-S, Kawabe T, Ono M, Aoki S, Sumizawa T, Furukawa T, Uchiumi T, Wada M, Kuwano M and Akiyama SI . (1999b). Mol. Pharmacol., 56, 1219–1228.

  19. Chen Z-S, Lee K and Kruh GD . (2001). J. Biol. Chem., 276, 33747–33754.

  20. Chen Z-S, Lee K, Walther S, Blanchard Raftogianis R, Kuwano M, Zeng H and Kruh GD . (2002). Cancer Res., 62, 3144–3150.

  21. Chen ZS, Hopper-Borge E, Belinsky MG, Shchaveleva I, Kotova E and Kruh GD . (2003a). Mol. Pharmacol., 63, 351–358.

  22. Chen Z-S, Robey RW, Belinsky MG, Shchaveleva I, Ren XQ, Sugimoto Y, Ross DD and Kruh GD . (2003b). Cancer Res., 63, 4048–4059.

  23. Chu XY, Kato Y, Niinuma K, Sudo KI, Hakusui H and Sugiyama Y . (1997). J. Pharmacol. Exp. Ther., 281, 304–314.

  24. Ciaccio PJ, Shen H, Kruh GD and Tew KD . (1996). Biochem. Biophys. Res. Commun., 222, 111–115.

  25. Cole SP, Bhardwaj G, Gerlach JH, Mackie JE, Grant CE, Almquist KC, Stewart AJ, Kurz EU, Duncan AM and Deeley RG . (1992). Science, 258, 1650–1654.

  26. Cole SP, Sparks KE, Fraser K, Loe DW, Grant CE, Wilson GM and Deeley RG . (1994). Cancer Res., 54, 5902–5910.

  27. Cui Y, Konig J, Buchholz JK, Spring H, Leier I and Keppler D . (1999). Mol. Pharmacol., 55, 929–937.

  28. Daoud R, Desneves J, Deady LW, Tilley L, Scheper RJ, Gros P and Georges E . (2000a). Biochemistry, 39, 6094–6102.

  29. Daoud R, Julien M, Gros P and Georges E . (2001). J. Biol. Chem., 276, 12324–12330.

  30. Daoud R, Kast C, Gros P and Georges E . (2000b). Biochemistry, 39, 15344–15352.

  31. Dietrich CG, de Waart DR, Ottenhoff R, Bootsma AH, van Gennip AH and Elferink RP . (2001). Carcinogenesis, 22, 805–811.

  32. Donner MG and Keppler D . (2001). Hepatology, 34, 351–359.

  33. Dubin IN and Johnson FB . (1954). Medicine, 33, 155–197.

  34. Elferink RP, Ottenhoff R, Liefting W, de Haan J and Jansen PL . (1989). J. Clin. Invest., 84, 476–483.

  35. Evers R, de Haas M, Sparidans R, Beijnen J, Wielinga PR, Lankelma J and Borst P . (2000). Br. J. Cancer, 83, 375–383.

  36. Evers R, Zaman GJ, van Deemter L, Jansen H, Calafat J, Oomen LC, Oude Elferink RP, Borst P and Schinkel AH . (1996). J. Clin. Invest., 97, 1211–1218.

  37. Fernandes DJ, Danks MK and Beck WT . (1990). Biochemistry, 29, 4235–4416.

  38. Flens MJ, Zaman GJ, van der Valk P, Izquierdo MA, Schroeijers AB, Scheffer GL, van der Groep P, de Haas M, Meijer CJ and Scheper RJ . (1996). Am. J. Pathol., 148, 1237–1247.

  39. Gao M, Yamazaki M, Loe DW, Westlake CJ, Grant CE, Cole SP and Deeley RG . (1998). J. Biol. Chem., 273, 10733–10740.

  40. Gerloff T, Stieger B, Hagenbuch B, Madon J, Landmann L, Roth J, Hofmann AF and Meier PJ . (1998). J. Biol. Chem., 273, 10046–10050.

  41. Guo Y, Chen Z-S, Kotova E, Hoppper-Borge E and Kruh GD . (2003). J. Biol. Chem., 278, 29509–29514.

  42. Gupta E, Safa AR, Wang X and Ratain MJ . (1996). Cancer Res., 56, 1309–1314.

  43. Haimeur A, Deeley RG and Cole SP . (2002). J. Biol. Chem., 277, 41326–41333.

  44. Harris MJ, Kuwano M, Webb M and Board PG . (2001). J. Biol. Chem., 276, 20876–20881.

  45. Hegedus T, Sessler T, Scott R, Thelin W, Bakos E, Varadi A, Szabo K, Homolya L, Milgram SL and Sarkadi B . (2003). Biochem. Biophys. Res. Commun., 302, 454–461.

  46. Hinoshita E, Uchiumi T, Taguchi K, Kinukawa N, Tsuneyoshi M, Maehara Y, Sugimachi K and Kuwano M . (2000). Clin. Cancer Res., 6, 2401–2407.

  47. Hipfner DR, Almquist KC, Leslie EM, Gerlach JH, Grant CE, Deeley RG and Cole SP . (1997). J. Biol. Chem., 272, 23623–23630.

  48. Hirohashi T, Suzuki H, Ito K, Ogawa K, Kume K, Shimizu T and Sugiyama Y . (1998). Mol. Pharmacol., 53, 1068–1075.

  49. Hirohashi T, Suzuki H and Sugiyama Y . (1999). J. Biol. Chem., 274, 15181–15185.

  50. Hirohashi T, Suzuki H, Takikawa H and Sugiyama Y . (2000). J. Biol. Chem., 275, 2905–2910.

  51. Hirrlinger J, Konig J, Keppler D, Lindenau J, Schulz JB and Dringen R . (2001). J. Neurochem., 76, 627–636.

  52. Hooijberg JH, Broxterman HJ, Kool M, Assaraf YG, Peters GJ, Noordhuis P, Scheper RJ, Borst P, Pinedo HM and Jansen G . (1999). Cancer Res., 59, 2532–2535.

  53. Hooijberg JH, Peters GJ, Assaraf YG, Kathmann I, Priest DG, Bunni MA, Veerman AJ, Scheffer GL, Kaspers GJ and Jansen G . (2003). Biochem. Pharmacol., 65, 765–771.

  54. Hopper E, Belinsky MG, Zeng H, Tosolini A, Testa JR and Kruh GD . (2001). Cancer Lett., 162, 181–191.

  55. Ilias A, Urban Z, Seidl TL, Le Saux O, Sinko E, Boyd CD, Sarkadi B and Varadi A . (2002). J. Biol. Chem., 277, 16860–16867.

  56. Inokuchi A, Hinoshita E, Iwamoto Y, Kohno K, Kuwano M and Uchiumi T . (2001). J. Biol. Chem., 4, 4.

  57. Ishikawa T . (1992). Trends Biochem. Sci., 17, 463–468.

  58. Ishikawa T and Ali-Osman F . (1993). J. Biol. Chem., 268, 20116–20125.

  59. Ito K, Oleschuk CJ, Westlake C, Vasa MZ, Deeley RG and Cole SP . (2001a). J. Biol. Chem., 276, 38108–38114.

  60. Ito K, Olsen SL, Qiu W, Deeley RG and Cole SP . (2001b). J. Biol. Chem., 276, 15616–15624.

  61. Ito K, Suzuki H and Sugiyama Y . (2001c). Mol. Pharmacol., 59, 1077–1085.

  62. Ito K, Suzuki H and Sugiyama Y . (2001d). Am. J. Physiol. Gastrointest. Liver Physiol., 281, G1034–G1043.

  63. Jansen PL, Peters WH and Lamers WH . (1985). Hepatology, 5, 573–579.

  64. Jedlitschky G, Burchell B and Keppler D . (2000). J. Biol. Chem., 275, 30069–30074.

  65. Jedlitschky G, Leier I, Buchholz U, Barnouin K, Kurz G and Keppler D . (1996). Cancer Res., 56, 988–994.

  66. Jedlitschky G, Leier I, Buchholz U, Hummel-Eisenbeiss J, Burchell B and Keppler D . (1997). Biochem. J., 327 (Part 1), 305–310.

  67. Johnson DR, Finch RA, Lin ZP, Zeiss CJ and Sartorelli AC . (2001). Cancer Res., 61, 1469–1476.

  68. Kao H, Huang J and Chang M . (2002). Gene, 286, 299–306.

  69. Kartenbeck J, Leuschner U, Mayer R and Keppler D . (1996). Hepatology, 23, 1061–1066.

  70. Karwatsky J, Daoud R, Cai J, Gros P and Georges E . (2003). Biochemistry, 42, 3286–3294.

  71. Kast C and Gros P . (1997). J. Biol. Chem., 272, 26479–26487.

  72. Kawabe T, Chen Z-S, Wada M, Uchiumi T, Ono M, Akiyama S and Kuwano M . (1999). FEBS Lett., 456, 327–331.

  73. Kitamura T, Jansen P, Hardenbrook C, Kamimoto Y, Gatmaitan Z and Arias IM . (1990). Proc. Natl. Acad. Sci. USA, 87, 3557–3561.

  74. Kiuchi Y, Suzuki H, Hirohashi T, Tyson CA and Sugiyama Y . (1998). FEBS Lett., 433, 149–152.

  75. Kocher O, Comella N, Gilchrist A, Pal R, Tognazzi K, Brown LF and Knoll JH . (1999). Lab. Invest., 79, 1161–1170.

  76. Koike K, Kawabe T, Tanaka T, Toh S, Uchiumi T, Wada M, Akiyama S, Ono M and Kuwano M . (1997). Cancer Res., 57, 5475–5479.

  77. Konno T, Ebihara T, Hisaeda K, Uchiumi T, Nakamura T, Shirakusa T, Kuwano M and Wada M . (2003). J. Biol. Chem., 278, 22908–22917.

  78. Kool M, van der Linden M, de Haas M, Baas F and Borst P . (1999a). Cancer Res., 59, 175–182.

  79. Kool M, van der Linden M, de Haas M, Scheffer GL, de Vree JM, Smith AJ, Jansen G, Peters GJ, Ponne N, Scheper RJ, Elferink RP, Baas F and Borst P . (1999b). Proc. Natl. Acad. Sci. USA, 96, 6914–6919.

  80. Krishnamachary N and Center MS . (1993). Cancer Res., 53, 3658–3661.

  81. Kruh GD, Chan A, Myers K, Gaughan K, Miki T and Aaronson SA . (1994). Cancer Res., 54, 1649–1652.

  82. Kruh GD, Gaughan KT, Godwin A and Chan A . (1995). J. Natl. Cancer Inst., 87, 1256–1258.

  83. Kruh GD, Zeng H, Rea PA, Liu G, Chen Z-S, Lee K and Belinsky MG . (2001). J. Bioenerg. Biomembr., 33, 493–501.

  84. Kuo MT, Bao JJ, Curley SA, Ikeguchi M, Johnston DA and Ishikawa T . (1996). Cancer Res., 56, 3642–3644.

  85. Lai L and Tan TM . (2002). Biochem. J., 361, 497–503.

  86. Le Saux O, Urban Z, Tschuch C, Csiszar K, Bacchelli B, Quaglino D, Pasquali-Ronchetti I, Pope FM, Richards A, Terry S, Bercovitch L, de Paepe A and Boyd CD . (2000). Nat. Genet., 25, 223–227.

  87. Lee K, Belinsky MG, Bell DW, Testa JR and Kruh GD . (1998). Cancer Res., 58, 2741–2747.

  88. Lee K, Klein-Szanto AJ and Kruh GD . (2000). J. Natl. Cancer Inst., 92, 1934–1940.

  89. Legare D, Richard D, Mukhopadhyay R, Stierhof YD, Rosen BP, Haimeur A, Papadopoulou B and Ouellette M . (2001). J. Biol. Chem., 276, 26301–26307.

  90. Leier I, Jedlitschky G, Buchholz U, Center M, Cole SP, Deeley RG and Keppler D . (1996a). Biochem. J., 314 (Part 2), 433–437.

  91. Leier I, Jedlitschky G, Buchholz U, Cole SP, Deeley RG and Keppler D . (1994). J. Biol. Chem., 269, 27807–27810.

  92. Leier I, Jedlitschky G, Buchler M, Buchholz U, Brom M and Keppler D . (1996b). Eur. J. Biochem., 242, 201–205.

  93. Leslie EM, Deeley RG and Cole SP . (2003a). Drug Metab. Dispos., 31, 11–15.

  94. Leslie EM, Ito K, Upadhyaya P, Hecht SS, Deeley RG and Cole SP . (2001). J. Biol. Chem., 276, 27846–27854.

  95. Leslie EM, Letourneau IJ, Deeley RG and Cole SP . (2003b). Biochemistry, 42, 5214–5224.

  96. Li ZS, Lu YP, Zhen RG, Szczypka M, Thiele DJ and Rea PA . (1997). Proc. Natl. Acad. Sci. USA, 94, 42–47.

  97. Lin ZP, Johnson DR, Finch RA, Belinsky MG, Kruh GD and Sartorelli AC . (2002). Mol. Cancer Ther., 1, 1105–1114.

  98. Loe DW, Almquist KC, Cole SP and Deeley RG . (1996a). J. Biol. Chem., 271, 9683–9689.

  99. Loe DW, Almquist KC, Deeley RG and Cole SP . (1996b). J. Biol. Chem., 271, 9675–9682.

  100. Loe DW, Deeley RG and Cole SP . (1998). Cancer Res., 58, 5130–5136.

  101. Loe DW, Deeley RG and Cole SP . (2000). J. Pharmacol. Exp. Ther., 293, 530–538.

  102. Lorico A, Bertola A, Baum C, Fodstad O and Rappa G . (2002). Biochem. Biophys. Res. Commun., 291, 617–622.

  103. Lorico A, Rappa G, Finch RA, Yang D, Flavell RA and Sartorelli AC . (1997). Cancer Res., 57, 5238–5242.

  104. Madon J, Hagenbuch B, Landmann L, Meier PJ and Stieger B . (2000). Mol. Pharmacol., 57, 634–641.

  105. Manciu L, Chang XB, Buyse F, Hou YX, Gustot A, Riordan JR and Ruysschaert JM . (2003). J. Biol. Chem., 278, 3347–3356.

  106. Mao Q, Qiu W, Weigl KE, Lander PA, Tabas LB, Shepard RL, Dantzig AH, Deeley RG and Cole SP . (2002). J. Biol. Chem., 277, 28690–28699.

  107. Marquardt D, McCrone S and Center MS . (1990). Cancer Res., 50, 1426–1430.

  108. Masuda M, I'izuka Y, Yamazaki M, Nishigaki R, Kato Y, Ni'inuma K, Suzuki H and Sugiyama Y . (1997). Cancer Res., 57, 3506–3510.

  109. McAleer MA, Breen MA, White NL and Matthews N . (1999). J. Biol. Chem., 274, 23541–23548.

  110. McGrath T and Center MS . (1987). Biochem. Biophys. Res. Commun., 145, 1171–1176.

  111. McGrath T, Latoud C, Arnold ST, Safa AR, Felsted RL and Center MS . (1989). Biochem. Pharmacol., 38, 3611–3619.

  112. Miller DS, Nobmann SN, Gutmann H, Toeroek M, Drewe J and Fricker G . (2000). Mol. Pharmacol., 58, 1357–1367.

  113. Morikawa A, Goto Y, Suzuki H, Hirohashi T and Sugiyama Y . (2000). Pharm. Res., 17, 546–552.

  114. Morrow CS, Smitherman PK, Diah SK, Schneider E and Townsend AJ . (1998). J. Biol. Chem., 273, 20114–20120.

  115. Mottino AD, Hoffman T, Jennes L and Vore M . (2000). J. Pharmacol. Exp. Ther., 293, 717–723.

  116. Moyer BD, Denton J, Karlson KH, Reynolds D, Wang S, Mickle JE, Milewski M, Cutting GR, Guggino WB, Li M and Stanton BA . (1999). J. Clin. Invest., 104, 1353–1361.

  117. Nies AT, Konig J, Cui Y, Brom M, Spring H and Keppler D . (2002a). Eur. J. Biochem., 269, 1866–1876.

  118. Nies AT, Konig J, Pfannschmidt M, Klar E, Hofmann WJ and Keppler D . (2001). Int. J. Cancer, 94, 492–499.

  119. Nies AT, Spring H, Thon WF, Keppler D and Jedlitschky G . (2002b). J. Urol., 167, 2271–2275.

  120. O'Brien M, Kruh GD and Tew KD . (2000). J. Pharmacol. Exp. Ther., 294, 480–487.

  121. Ohishi Y, Oda Y, Uchiumi T, Kobayashi H, Hirakawa T, Miyamoto S, Kinukawa N, Nakano H, Kuwano M and Tsuneyoshi M . (2002). Clin. Cancer Res., 8, 3767–3775.

  122. Paulusma CC, Bosma PJ, Zaman GJ, Bakker CT, Otter M, Scheffer GL, Scheper RJ, Borst P and Oude Elferink RP . (1996). Science, 271, 1126–1128.

  123. Paumi CM, Ledford BG, Smitherman PK, Townsend AJ and Morrow CS . (2001). J. Biol. Chem., 276, 7952–7956.

  124. Plasschaert SL, Vellenga E, de Bont ES, van der Kolk DM, Veerman AJ, Sluiter WJ, Daenen SM, de Vries EG and Kamps WA . (2003). Leukemia Lymphoma, 44, 85–95.

  125. Potschka H, Fedrowitz M and Loscher W . (2003). J. Pharmacol. Exp. Ther.

  126. Qian YM, Grant CE, Westlake CJ, Zhang DW, Lander PA, Shepard RL, Dantzig AH, Cole SP and Deeley RG . (2002). J. Biol. Chem., 277, 35225–35231.

  127. Qian YM, Song WC, Cui H, Cole SP and Deeley RG . (2001). J. Biol. Chem., 276, 6404–6411.

  128. Randolph GJ . (2001). Semin. Immunol., 13, 267–274.

  129. Rao VV, Dahlheimer JL, Bardgett ME, Snyder AZ, Finch RA, Sartorelli AC and Piwnica-Worms D . (1999). Proc. Natl. Acad. Sci. USA, 96, 3900–3905.

  130. Rappa G, Lorico A, Flavell RA and Sartorelli AC . (1997). Cancer Res., 57, 5232–5237.

  131. Ren XQ, Furukawa T, Aoki S, Nakajima T, Sumizawa T, Haraguchi M, Chen ZS, Kobayashi M and Akiyama S . (2001). J. Biol. Chem., 276, 23197–23206.

  132. Ren XQ, Furukawa T, Aoki S, Sumizawa T, Haraguchi M, Nakajima Y, Ikeda R, Kobayashi M and Akiyama S . (2002). Biochemistry, 41, 14132–14140.

  133. Ringpfeil F, Lebwohl MG, Christiano AM and Uitto J . (2000). Proc. Natl. Acad. Sci. USA, 97, 6001–6006.

  134. Robbiani DF, Finch RA, Jager D, Muller WA, Sartorelli AC and Randolph GJ . (2000). Cell, 103, 757–768.

  135. Robbins BL, Connelly MC, Marshall DR, Srinivas RV and Fridland A . (1995). Mol. Pharmacol., 47, 391–397.

  136. Rosenberg MF, Mao Q, Holzenburg A, Ford RC, Deeley RG and Cole SP . (2001). J. Biol. Chem., 276, 16076–16082.

  137. Rost D, Mahner S, Sugiyama Y and Stremmel W . (2002). Am. J. Physiol. Gastrointest. Liver Physiol., 282, G720–6.

  138. Ryu S, Kawabe T, Nada S and Yamaguchi A . (2000). J. Biol. Chem., 275, 39617–39624.

  139. Sakamoto H, Hara H, Hirano K and Adachi T . (1999). Cancer Lett., 135, 113–119.

  140. Sanghani PC and Moran RG . (2000). Protein Expres. Purif., 18, 36–45.

  141. Sauerbrey A, Voigt A, Wittig S, Hafer R and Zintl F . (2002). Leukemia Lymphoma, 43, 875–879.

  142. Schaub TP, Kartenbeck J, Konig J, Vogel O, Witzgall R, Kriz W and Keppler D . (1997). J. Am. Soc. Nephrol., 8, 1213–1221.

  143. Scheffer GL, Hu X, Pijnenborg AC, Wijnholds J, Bergen AA and Scheper RJ . (2002a). Lab. Invest., 82, 515–518.

  144. Scheffer GL, Kool M, de Haas M, de Vree JM, Pijnenborg AC, Bosman DK, Elferink RP, van der Valk P, Borst P and Scheper RJ . (2002b). Lab. Invest., 82, 193–201.

  145. Schneider E, Yamazaki H, Sinha BK and Cowan KH . (1995). Br. J. Cancer, 71, 738–743.

  146. Schuetz JD, Connelly MC, Sun D, Paibir SG, Flynn PM, Srinivas RV, Kumar A and Fridland A . (1999). Nat. Med., 5, 1048–1051.

  147. Schultz MJ, Wijnholds J, Peppelenbosch MP, Vervoordeldonk MJ, Speelman P, van Deventer SJ, Borst P and van der Poll T . (2001). J. Immunol., 166, 4059–4064.

  148. Soderling SH and Beavo JA . (2000). Curr. Opin. Cell Biol., 12, 174–179.

  149. Soroka CJ, Lee JM, Azzaroli F and Boyer JL . (2001). Hepatology, 33, 783–791.

  150. St-Pierre MV, Serrano MA, Macias RI, Dubs U, Hoechli M, Lauper U, Meier PJ and Marin JJ . (2000). Am. J. Physiol. Regul. Integr. Comp. Physiol., 279, R1495–503.

  151. Steinbach D, Lengemann J, Voigt A, Hermann J, Zintl F and Sauerbrey A . (2003). Clin. Cancer Res., 9, 1083–1086.

  152. Suzuki H and Sugiyama Y . (2002). Adv. Drug Deliv. Rev., 54, 1311–1331.

  153. Taguchi Y, Yoshida A, Takada Y, Komano T and Ueda K . (1997). FEBS Lett., 401, 11–14.

  154. Tammur J, Prades C, Arnould I, Rzhetsky A, Hutchinson A, Adachi M, Schuetz JD, Swoboda KJ, Ptacek LJ, Rosier M, Dean M and Allikmets R . (2001). Gene, 273, 89–96.

  155. Taniguchi K, Wada M, Kohno K, Nakamura T, Kawabe T, Kawakami M, Kagotani K, Okumura K, Akiyama S and Kuwano M . (1996). Cancer Res., 56, 4124–4129.

  156. Uchiumi T, Hinoshita E, Haga S, Nakamura T, Tanaka T, Toh S, Furukawa M, Kawabe T, Wada M, Kagotani K, Okumura K, Kohno K, Akiyama S and Kuwano M . (1998). Biochem. Biophys. Res. Commun., 252, 103–110.

  157. Uitto J, Pulkkinen L and Ringpfeil F . (2001). Trends Mol. Med., 7, 13–17.

  158. Van Aubel RA, Koenderink JB, Peters JG, Van Os CH and Russel FG . (1999). Mol. Pharmacol., 56, 714–719.

  159. van Aubel RA, Smeets PH, Peters JG, Bindels RJ and Russel FG . (2002). J. Am. Soc. Nephrol., 13, 595–603.

  160. van den Heuvel-Eibrink MM, Wiemer EA, Prins A, Meijerink JP, Vossebeld PJ, van der Holt B, Pieters R and Sonneveld P . (2002). Leukemia, 16, 833–839.

  161. Vernhet L, Allain N, Payen L, Anger JP, Guillouzo A and Fardel O . (2001). Biochem. Pharmacol., 61, 1387–1391.

  162. Vernhet L, Courtois A, Allain N, Payen L, Anger JP, Guillouzo A and Fardel O . (1999). FEBS Lett., 443, 321–325.

  163. Versantvoort CH, Broxterman HJ, Bagrij T, Scheper RJ and Twentyman PR . (1995). Br. J. Cancer, 72, 82–89.

  164. Volk EL, Farley KM, Wu Y, Li F, Robey RW and Schneider E . (2002). Cancer Res., 62, 5035–5040.

  165. Volk EL, Rohde K, Rhee M, McGuire JJ, Doyle LA, Ross DD and Schneider E . (2000). Cancer Res., 60, 3514–3521.

  166. Wielinga PR, Reid G, Challa EE, van der Heijden I, van Deemter L, de Haas M, Mol C, Kuil AJ, Groeneveld E, Schuetz JD, Brouwer C, De Abreu RA, Wijnholds J, Beijnen JH and Borst P . (2002). Mol. Pharmacol., 62, 1321–1331.

  167. Wielinga PR, Van Der Heijden I, Reid G, Beijnen JH, Wijnholds J and Borst P . (2003). J. Biol. Chem., 278, 17664–17671.

  168. Wijnholds J, deLange EC, Scheffer GL, van den Berg DJ, Mol CA, van der Valk M, Schinkel AH, Scheper RJ, Breimer DD and Borst P . (2000a). J. Clin. Invest., 105, 279–285.

  169. Wijnholds J, Evers R, van Leusden MR, Mol CA, Zaman GJ, Mayer U, Beijnen JH, van der Valk M, Krimpenfort P and Borst P . (1997). Nat. Med., 3, 1275–1279.

  170. Wijnholds J, Mol CA, van Deemter L, de Haas M, Scheffer GL, Baas F, Beijnen JH, Scheper RJ, Hatse S, De Clercq E, Balzarini J and Borst P . (2000b). Proc. Natl. Acad. Sci. USA, 97, 7476–7481.

  171. Yabuuchi H, Shimizu H, Takayanagi S and Ishikawa T . (2001). Biochem. Biophys. Res. Commun., 288, 933–939.

  172. Yang Y, Chen Q and Zhang JT . (2002). J. Biol. Chem., 277, 44268–44277.

  173. Zaman GJ, Flens MJ, van Leusden MR, de Haas M, Mulder HS, Lankelma J, Pinedo HM, Scheper RJ, Baas F, Broxterman HJ and Borst P . (1994). Proc. Natl. Acad. Sci. USA, 91, 8822–8826.

  174. Zaman GJ, Lankelma J, van Tellingen O, Beijnen J, Dekker H, Paulusma C, Oude Elferink RP, Baas F and Borst P . (1995). Proc. Natl. Acad. Sci. USA, 92, 7690–7694.

  175. Zelcer N, Huisman MT, Reid G, Wielinga PR, Breedveld P, Kuil A, Knipscheer P, Schellens JH, Schinkel AH and Borst P . (2003a). J. Biol. Chem., 278, 23538–23544.

  176. Zelcer N, Reid G, Wielinga P, Kuil A, Van Der Heijden I, Schuetz JD and Borst P . (2003b). Biochem. J., 371, 361–367.

  177. Zelcer N, Saeki T, Reid G, Beijnen JH and Borst P . (2001). J. Biol. Chem., 276, 46400–46407.

  178. Zeng H, Bain LJ, Belinsky MG and Kruh GD . (1999). Cancer Res., 59, 5964–5967.

  179. Zeng H, Chen Z-S, Belinsky MG, Rea PA and Kruh GD . (2001). Cancer Res., 61, 7225–7232.

  180. Zeng H, Liu G, Rea PA and Kruh GD . (2000). Cancer Res., 60, 4779–4784.

  181. Zhang DW, Cole SP and Deeley RG . (2001a). J. Biol. Chem., 276, 34966–34974.

  182. Zhang DW, Cole SP and Deeley RG . (2001b). J. Biol. Chem., 276, 13231–13239.

  183. Zhang DW, Cole SP and Deeley RG . (2002). J. Biol. Chem., 277, 20934–20941.

  184. Zurita AJ, Diestra JE, Condom E, Garcia Del Muro X, Scheffer GL, Scheper RJ, Perez J, Germa-Lluch JR and Izquierdo MA . (2003). Br. J. Cancer, 88, 879–886.

Download references


This work was supported in part by National Institutes of health Grants CA73728 (to GDK) and CA06927 to the Fox Chase Cancer Center and by an appropriation from the Commonwealth of Pennsylvania.

Author information

Correspondence to Gary D Kruh.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kruh, G., Belinsky, M. The MRP family of drug efflux pumps. Oncogene 22, 7537–7552 (2003) doi:10.1038/sj.onc.1206953

Download citation


  • ABC transporter
  • efflux
  • resistance

Further reading